Building an AI-First Hyperscaler: Lessons Learnt from E2E Networks' Chief Technology Officer

July 6, 2023

Mr Imran is a grounded man. Speak to him about anything tech - and he will instantly highlight how India is coming to the foreground in technology more than ever before. As the CTO of E2E Networks, he has had his fair share of challenges and triumphs, leading their tech team for over a decade from the hitech city of Hyderabad. 

We had a tête-à-tête with him when he recently visited Delhi for a conference. 

Can you talk a little bit about what E2E’s vision is - about growing and scaling a Swadeshi AI-First Hyperscaler - and your journey through this process? 

E2E’s vision is to build a complete suite of tools to make it easier and more efficient to deploy modern-day AI workloads. There is also a strong focus on leveraging open source technologies both for building the tech stack as well as for making available open source models which can be further trained and modified according to the requirements of customers. Eventually the days of manually setting up infrastructure, monitoring it, and scaling it manually as an AI company is a complete distraction from the core requirement of building AI solutions. E2E aims to be that platform which takes into account all of these underlying requirements and offers a one-stop solution to businesses to train and deploy their AI models.

When did you join E2E? How has your journey been?

I joined E2E at the time of its inception, around 2010, and it's been a fulfilling journey to say the least. We started with building VMs (Virtual Machines) manually; then we scaled. The transition and the journey from an IAAS (Infrastructure As a Service) offering to building scalable products like storage, networks and so on, and then further into AI solutions, has been a tremendous challenge and a really enjoyable one. 

Tech is the backbone of E2E. Tell us your experience of scaling a tech team as the company grew. 

At E2E, we believe that our people are the core strength of our company. If you see across the board, we have team members who have spent more than 8 years in the company, right from Tech to Sales to Finance. So, naturally, the way we scaled up was to empower our best performers with more responsibility and then building teams around them. 

In tech, we have benefited from our extensive experience in open source technologies and are able to build standardised solutions. Once you are in complete control of your technology stack using open source, it's easier to scale it as well. 

How much do you rely on open-source solutions, and why?

It's an intentional choice to use standard-based open source solutions simply because of the no-strings attached availability and licensing. That is one of the core reasons behind our ability to scale. Secondly, we actually pay for enterprise support for a lot of our open source usage - not just for support but as a reasonable consumer of open source, we want to show our solidarity with the companies building open source products. Thirdly, it's easier to scale an open source product for availability, resiliency and scale as there is a vibrant community available to bounce off ideas with. 

How do you build an organization that can keep up with the pace of change that’s currently happening in AI?

Eventually it is people, people, and people. AI is a rapidly evolving paradigm and we have now built teams who have worked on AI for a period of 4-5 years. As again, we leverage open source offerings like Kubeflow and other products, which are pretty much setting the standard on how to do AI as a service. 

The selection of the tech stack itself has a major advantage for us. Add to this our ability to do massive performance on infrastructure and network, and we are able to deliver high-performant AI solutions to our customers.

What kinds of technology problems do you think will drive the next wave of innovation?

The ability to onboard a huge number of users onto AI is where the biggest challenge and opportunity lies. The question is, how do we help data scientists build models at scale (read PBs of data), deploy them for scale (let’s say, a publicly available API), and monetize those offerings. This is going to drive the next wave of innovation. 

The current buzz around GPT and generative AI will drive a lot of users to AI and there will be a significant opportunity to streamline this process. 

What drives clients to E2E?

Although a principal amount of discovery happens because of our price to performance ratio, our customers have started appreciating our advantage of being a strong tech company leveraging open source solutions to build services. We are also seeing significant attention to our GPU products with A100, A30, L4 and other cards and the soon to be launched latest gen cards as well. 

How do you go about getting employees to begin to think a little differently about what the technologies are capable of, raising that technological imagination in the team?

A lot of the foundational changes in terms of how teams think has to do with transparency of the company’s vision and strategy. We have always communicated our vision and core strategy to all stakeholders internally and that gets the teams excited. At E2E, we don't build products in silos. It's common for a sales or a finance person to join a tech demo and give suggestions based on what they hear as feedback from our customers. 

Of course, we invest in employee training programs, encouraging employees to attend conferences etc. to widen their technological worldview. In a nutshell, transparency breeds trust; trust fuels innovation; and innovation delivers customer value. 

Could you relate some experiences with your team that have been special?

We have had deployments that have stretched on for days. To see my team members self-organise themselves to guarantee 24/7 coverage for both deployment and support was a satisfying experience on having built a team which is aligned with the requirements of the company and not just driven by me as the manager. 

E2E is one of the largest swadeshi AI-first Hyperscaler platforms. What have been the 3 biggest challenges of building from India for the world?

The first one would be access to capital to build an AI-first platform as the existing hyperscalers are all able to deploy millions or billions of dollars in a short span of time. 

Secondly, a lot of nations have their own privacy laws like the GDPR (General Data Protection Regulation) in Europe and there are similar laws in the US as well which restrict how much compliance you can do. Being an Indian company domiciled in India, we of course have to adhere to Indian laws above all else. A data protection framework which creates a level-playing field for Indian companies would be a good starting point. 

Thirdly, we need to start training our students/professionals in generative AI so they can build products and models rather than just working in service companies. 

Can you relate a couple of key incidents from your career at E2E that have transformed you as a leader? 

Covid had a dramatic impact on the ability to lead a team simply because of having to reimagine everything you are used to. Secondly, the transition to doing more products rather than simple IAAS meant we had to restructure the teams to be able to deliver them. 

Can you tell us what you look for in an engineer before they are hired in E2E? 

The ability to find solutions and demonstrate actual proficiency in code/ops rather than simply subject matter expertise is a key differentiator. Other things include the ability to ask smart questions, communication skills, technical skill upgrade history, and so on. 

Finally, can you talk a bit about the future of the Indian tech ecosystem, and where you see it headed? 

India is going to be generating massive amounts of data and the tech ecosystem would be transitioning to build products with these large data sets. The fundamentals of the industry will change as more AI penetration happens in academia and availability of trained machine learning engineers happens at scale. That would be an epochal moment for the Indian tech industry - to leverage that talent and build more products and services suited for a global landscape. I see us transitioning out of the services model and stepping directly into products. 

Do you agree with Sam Altman’s comment on India’s chances of building a foundational AI company? 

AI requires massive amounts of funding simply because of the GPU compute required to do training on the LLMs. Sam isn't far off the mark when he says that you cannot build a foundational AI company for 10 million dollars. The real momentum in building such companies would come in with increased funding to invest over a longer period of time for building foundational AI/AGI. 

We have everything it takes in terms of tech talent, availability of infrastructure, etc. to be able to build such companies. The only thing we require is sustained funding over a longer cycle to leapfrog into a foundational AI company. 

Latest Blogs
This is a decorative image for: A Complete Guide To Customer Acquisition For Startups
October 18, 2022

A Complete Guide To Customer Acquisition For Startups

Any business is enlivened by its customers. Therefore, a strategy to constantly bring in new clients is an ongoing requirement. In this regard, having a proper customer acquisition strategy can be of great importance.

So, if you are just starting your business, or planning to expand it, read on to learn more about this concept.

The problem with customer acquisition

As an organization, when working in a diverse and competitive market like India, you need to have a well-defined customer acquisition strategy to attain success. However, this is where most startups struggle. Now, you may have a great product or service, but if you are not in the right place targeting the right demographic, you are not likely to get the results you want.

To resolve this, typically, companies invest, but if that is not channelized properly, it will be futile.

So, the best way out of this dilemma is to have a clear customer acquisition strategy in place.

How can you create the ideal customer acquisition strategy for your business?

  • Define what your goals are

You need to define your goals so that you can meet the revenue expectations you have for the current fiscal year. You need to find a value for the metrics –

  • MRR – Monthly recurring revenue, which tells you all the income that can be generated from all your income channels.
  • CLV – Customer lifetime value tells you how much a customer is willing to spend on your business during your mutual relationship duration.  
  • CAC – Customer acquisition costs, which tells how much your organization needs to spend to acquire customers constantly.
  • Churn rate – It tells you the rate at which customers stop doing business.

All these metrics tell you how well you will be able to grow your business and revenue.

  • Identify your ideal customers

You need to understand who your current customers are and who your target customers are. Once you are aware of your customer base, you can focus your energies in that direction and get the maximum sale of your products or services. You can also understand what your customers require through various analytics and markers and address them to leverage your products/services towards them.

  • Choose your channels for customer acquisition

How will you acquire customers who will eventually tell at what scale and at what rate you need to expand your business? You could market and sell your products on social media channels like Instagram, Facebook and YouTube, or invest in paid marketing like Google Ads. You need to develop a unique strategy for each of these channels. 

  • Communicate with your customers

If you know exactly what your customers have in mind, then you will be able to develop your customer strategy with a clear perspective in mind. You can do it through surveys or customer opinion forms, email contact forms, blog posts and social media posts. After that, you just need to measure the analytics, clearly understand the insights, and improve your strategy accordingly.

Combining these strategies with your long-term business plan will bring results. However, there will be challenges on the way, where you need to adapt as per the requirements to make the most of it. At the same time, introducing new technologies like AI and ML can also solve such issues easily. To learn more about the use of AI and ML and how they are transforming businesses, keep referring to the blog section of E2E Networks.

Reference Links

This is a decorative image for: Constructing 3D objects through Deep Learning
October 18, 2022

Image-based 3D Object Reconstruction State-of-the-Art and trends in the Deep Learning Era

3D reconstruction is one of the most complex issues of deep learning systems. There have been multiple types of research in this field, and almost everything has been tried on it — computer vision, computer graphics and machine learning, but to no avail. However, that has resulted in CNN or convolutional neural networks foraying into this field, which has yielded some success.

The Main Objective of the 3D Object Reconstruction

Developing this deep learning technology aims to infer the shape of 3D objects from 2D images. So, to conduct the experiment, you need the following:

  • Highly calibrated cameras that take a photograph of the image from various angles.
  • Large training datasets can predict the geometry of the object whose 3D image reconstruction needs to be done. These datasets can be collected from a database of images, or they can be collected and sampled from a video.

By using the apparatus and datasets, you will be able to proceed with the 3D reconstruction from 2D datasets.

State-of-the-art Technology Used by the Datasets for the Reconstruction of 3D Objects

The technology used for this purpose needs to stick to the following parameters:

  • Input

Training with the help of one or multiple RGB images, where the segmentation of the 3D ground truth needs to be done. It could be one image, multiple images or even a video stream.

The testing will also be done on the same parameters, which will also help to create a uniform, cluttered background, or both.

  • Output

The volumetric output will be done in both high and low resolution, and the surface output will be generated through parameterisation, template deformation and point cloud. Moreover, the direct and intermediate outputs will be calculated this way.

  • Network architecture used

The architecture used in training is 3D-VAE-GAN, which has an encoder and a decoder, with TL-Net and conditional GAN. At the same time, the testing architecture is 3D-VAE, which has an encoder and a decoder.

  • Training used

The degree of supervision used in 2D vs 3D supervision, weak supervision along with loss functions have to be included in this system. The training procedure is adversarial training with joint 2D and 3D embeddings. Also, the network architecture is extremely important for the speed and processing quality of the output images.

  • Practical applications and use cases

Volumetric representations and surface representations can do the reconstruction. Powerful computer systems need to be used for reconstruction.

Given below are some of the places where 3D Object Reconstruction Deep Learning Systems are used:

  • 3D reconstruction technology can be used in the Police Department for drawing the faces of criminals whose images have been procured from a crime site where their faces are not completely revealed.
  • It can be used for re-modelling ruins at ancient architectural sites. The rubble or the debris stubs of structures can be used to recreate the entire building structure and get an idea of how it looked in the past.
  • They can be used in plastic surgery where the organs, face, limbs or any other portion of the body has been damaged and needs to be rebuilt.
  • It can be used in airport security, where concealed shapes can be used for guessing whether a person is armed or is carrying explosives or not.
  • It can also help in completing DNA sequences.

So, if you are planning to implement this technology, then you can rent the required infrastructure from E2E Networks and avoid investing in it. And if you plan to learn more about such topics, then keep a tab on the blog section of the website

Reference Links

This is a decorative image for: Comprehensive Guide to Deep Q-Learning for Data Science Enthusiasts
October 18, 2022

A Comprehensive Guide To Deep Q-Learning For Data Science Enthusiasts

For all data science enthusiasts who would love to dig deep, we have composed a write-up about Q-Learning specifically for you all. Deep Q-Learning and Reinforcement learning (RL) are extremely popular these days. These two data science methodologies use Python libraries like TensorFlow 2 and openAI’s Gym environment.

So, read on to know more.

What is Deep Q-Learning?

Deep Q-Learning utilizes the principles of Q-learning, but instead of using the Q-table, it uses the neural network. The algorithm of deep Q-Learning uses the states as input and the optimal Q-value of every action possible as the output. The agent gathers and stores all the previous experiences in the memory of the trained tuple in the following order:

State> Next state> Action> Reward

The neural network training stability increases using a random batch of previous data by using the experience replay. Experience replay also means the previous experiences stocking, and the target network uses it for training and calculation of the Q-network and the predicted Q-Value. This neural network uses openAI Gym, which is provided by taxi-v3 environments.

Now, any understanding of Deep Q-Learning   is incomplete without talking about Reinforcement Learning.

What is Reinforcement Learning?

Reinforcement is a subsection of ML. This part of ML is related to the action in which an environmental agent participates in a reward-based system and uses Reinforcement Learning to maximize the rewards. Reinforcement Learning is a different technique from unsupervised learning or supervised learning because it does not require a supervised input/output pair. The number of corrections is also less, so it is a highly efficient technique.

Now, the understanding of reinforcement learning is incomplete without knowing about Markov Decision Process (MDP). MDP is involved with each state that has been presented in the results of the environment, derived from the state previously there. The information which composes both states is gathered and transferred to the decision process. The task of the chosen agent is to maximize the awards. The MDP optimizes the actions and helps construct the optimal policy.

For developing the MDP, you need to follow the Q-Learning Algorithm, which is an extremely important part of data science and machine learning.

What is Q-Learning Algorithm?

The process of Q-Learning is important for understanding the data from scratch. It involves defining the parameters, choosing the actions from the current state and also choosing the actions from the previous state and then developing a Q-table for maximizing the results or output rewards.

The 4 steps that are involved in Q-Learning:

  1. Initializing parameters – The RL (reinforcement learning) model learns the set of actions that the agent requires in the state, environment and time.
  2. Identifying current state – The model stores the prior records for optimal action definition for maximizing the results. For acting in the present state, the state needs to be identified and perform an action combination for it.
  3. Choosing the optimal action set and gaining the relevant experience – A Q-table is generated from the data with a set of specific states and actions, and the weight of this data is calculated for updating the Q-Table to the following step.
  4. Updating Q-table rewards and next state determination – After the relevant experience is gained and agents start getting environmental records. The reward amplitude helps to present the subsequent step.  

In case the Q-table size is huge, then the generation of the model is a time-consuming process. This situation requires Deep Q-learning.

Hopefully, this write-up has provided an outline of Deep Q-Learning and its related concepts. If you wish to learn more about such topics, then keep a tab on the blog section of the E2E Networks website.

Reference Links

This is a decorative image for: GAUDI: A Neural Architect for Immersive 3D Scene Generation
October 13, 2022

GAUDI: A Neural Architect for Immersive 3D Scene Generation

The evolution of artificial intelligence in the past decade has been staggering, and now the focus is shifting towards AI and ML systems to understand and generate 3D spaces. As a result, there has been extensive research on manipulating 3D generative models. In this regard, Apple’s AI and ML scientists have developed GAUDI, a method specifically for this job.

An introduction to GAUDI

The GAUDI 3D immersive technique founders named it after the famous architect Antoni Gaudi. This AI model takes the help of a camera pose decoder, which enables it to guess the possible camera angles of a scene. Hence, the decoder then makes it possible to predict the 3D canvas from almost every angle.

What does GAUDI do?

GAUDI can perform multiple functions –

  • The extensions of these generative models have a tremendous effect on ML and computer vision. Pragmatically, such models are highly useful. They are applied in model-based reinforcement learning and planning world models, SLAM is s, or 3D content creation.
  • Generative modelling for 3D objects has been used for generating scenes using graf, pigan, and gsn, which incorporate a GAN (Generative Adversarial Network). The generator codes radiance fields exclusively. Using the 3D space in the scene along with the camera pose generates the 3D image from that point. This point has a density scalar and RGB value for that specific point in 3D space. This can be done from a 2D camera view. It does this by imposing 3D datasets on those 2D shots. It isolates various objects and scenes and combines them to render a new scene altogether.
  • GAUDI also removes GANs pathologies like mode collapse and improved GAN.
  • GAUDI also uses this to train data on a canonical coordinate system. You can compare it by looking at the trajectory of the scenes.

How is GAUDI applied to the content?

The steps of application for GAUDI have been given below:

  • Each trajectory is created, which consists of a sequence of posed images (These images are from a 3D scene) encoded into a latent representation. This representation which has a radiance field or what we refer to as the 3D scene and the camera path is created in a disentangled way. The results are interpreted as free parameters. The problem is optimized by and formulation of a reconstruction objective.
  • This simple training process is then scaled to trajectories, thousands of them creating a large number of views. The model samples the radiance fields totally from the previous distribution that the model has learned.
  • The scenes are thus synthesized by interpolation within the hidden space.
  • The scaling of 3D scenes generates many scenes that contain thousands of images. During training, there is no issue related to canonical orientation or mode collapse.
  • A novel de-noising optimization technique is used to find hidden representations that collaborate in modelling the camera poses and the radiance field to create multiple datasets with state-of-the-art performance in generating 3D scenes by building a setup that uses images and text.

To conclude, GAUDI has more capabilities and can also be used for sampling various images and video datasets. Furthermore, this will make a foray into AR (augmented reality) and VR (virtual reality). With GAUDI in hand, the sky is only the limit in the field of media creation. So, if you enjoy reading about the latest development in the field of AI and ML, then keep a tab on the blog section of the E2E Networks website.

Reference Links

Build on the most powerful infrastructure cloud

A vector illustration of a tech city using latest cloud technologies & infrastructure